Effect of Wearable Health Technology on Patients Treated for Chronic Pain at Geisinger Health System: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Wearable Health Technology on Patients Treated for Chronic Pain at Geisinger Health System (NCT03299556), a registered phase not stated study. The primary outcome is Numerical Pain Score (NPS).

2. Clinical Question

This is a proof-of-concept question about a technology package used in a clinical pain-care setting, not a clean comparison of a single device feature. The wearable group is described as receiving WHT alongside MPP care. The historic group received MPP education without WHT, while the concurrent group came from a different pain-management program and received neither MPP education nor WHT. These groups consequently differ in more than wearable technology. The allocation method is not stated, and the record characterizes the arms as not being a like-for-like control comparison. Differences between groups, if reported elsewhere, would need to be interpreted with that structure in mind.

3. Trial Design

The registry describes a completed proof-of-concept study at Geisinger Health System. It lists an enrollment of 420 and a status of completed. The study start date is 2017-10-05, and the primary completion date is 2019-11-30. Patients in the WHT group were newly enrolled in the Geisinger MPP and consented to participate. The group description says recruitment would occur over 1 year, with 1 year of follow-up.

Three groups are described. WHT participants used the wearable technology package. The historic control consisted of MPP patients enrolled in the preceding year; they received the MPP educational program but no WHT as part of treatment. The concurrent control consisted of patients treated for chronic pain in the Geisinger MPM program over the same period as the WHT group; they did not receive the MPP educational program or WHT. Allocation is not stated in the registry record. The record explicitly flags that this was not a randomized comparison and that the arms were not a like-for-like control comparison.

The technology was intended to combine self-reported measures with passively collected information. Self-reported information included pain, pain-management therapies and medication use; passive collection covered activity and sleep. The brief summary also names physical function, depression, healthcare resource utilization and cost among outcomes of interest. The listed secondary outcome is the Oswestry Disability Index (ODI). The registry identifies six outcome measures, but the available record does not describe all of them in detail.

The primary endpoint was NPS at baseline and 12-month follow-up. The record does not report the analysis population, the handling of follow-up observations, or a statistical analysis plan. It also does not state how the groups were formed beyond their program and time-period descriptions. Those omissions constrain interpretation: the design information establishes what was intended to be compared, but not how any observed differences were analyzed or adjusted.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Numerical Pain Score (NPS), measured over Baseline, 12-month follow-up.

The primary endpoint is the Numerical Pain Score (NPS), a patient-reported pain rating on a scale from 0 to 10. The registry describes the anchors as no pain (0), mild pain (2), moderate pain (4), severe pain (6), very severe pain (8), and worst pain possible (10). The planned assessment points listed for this endpoint are baseline and 12-month follow-up. The registry specifies the mean as the summary measure and standard deviation as the dispersion measure.

A patient-reported pain score can represent the experience of pain as described by the person receiving care. Comparing baseline and follow-up measurements can, in general, describe how reported scores change over time. A group mean gives a compact summary of scores, while the standard deviation describes how spread out individual scores are around that mean. These summaries answer different questions: the mean concerns the average score, whereas the standard deviation concerns variability among the recorded scores.

General statistical caution is important for a bounded rating scale. A mean alone does not show the full distribution, including whether scores are concentrated, widely dispersed, or clustered near the scale limits. A difference in average score would also need a stated comparison, uncertainty estimate and analysis approach to convey its precision. The registry record does not report the primary NPS means, standard deviations, changes, confidence intervals or group comparisons. As a result, the endpoint definition is available, but its observed result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies the primary outcome as a continuous measure summarized by its mean and standard deviation. It does not report a statistical analysis plan, a test or model, an estimate of change, a confidence interval, or a p-value. Statistical concepts can still clarify what such quantities would mean in general, but they cannot fill in the missing trial results.

What does a mean describe?

In general, a mean is a measure of the center of a set of observations. For NPS, it would summarize the average recorded rating in a group at a specified assessment point. A mean can make group summaries easier to compare, but it does not reveal whether individual ratings are similar or whether the average is shaped by a smaller number of unusually high or low observations. The registry specifies a mean but does not list the group means or the underlying score distributions.

What does a standard deviation describe?

In general, a standard deviation describes the spread of observations around their mean. A smaller standard deviation indicates that scores are more concentrated around the group mean; a larger one indicates more dispersion. It is not the same as the uncertainty in the estimated mean, and it is not itself a measure of the size or precision of a between-group effect. The record specifies standard deviation as the dispersion measure, but does not report its values for NPS.

How would a confidence interval help?

In general, a confidence interval gives a range of values compatible with an estimate under the assumptions of the analysis. Its width conveys precision: a narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. For a group comparison, an interval around the estimated contrast would help readers assess both the plausible size and direction of the difference. The registry record does not report a confidence interval for the primary endpoint, so the precision of any contrast is not available.

What does a significance test answer?

In general, a significance test evaluates how compatible observed data are with a specified null hypothesis under the model and assumptions used. A p-value is not the probability that the null hypothesis is true, nor does it state the size or clinical importance of an effect. It is most informative when considered alongside the estimated effect, its confidence interval, the design and the analysis plan. The record does not report a primary-endpoint p-value or the test used. No statistical significance claim can be made from the endpoint definition alone.

Why does the comparison structure affect analysis?

General statistical explanation: random allocation, when used and implemented, can help balance measured and unmeasured characteristics on average across groups. Here, allocation is not stated, and the registry flags that the comparisons are not randomized and not like-for-like. The WHT and historic groups differ in enrollment period, while the concurrent group is from a different care program and does not receive MPP education. A raw comparison could therefore mix differences associated with technology exposure with differences related to program, period or patient composition. The registry record does not report adjustment methods or enough detail to assess how such factors were handled.

Repeated measurements raise additional analytic questions. General methods may compare follow-up scores while accounting for baseline, compare change from baseline, or model repeated observations; these approaches make different assumptions and answer related but distinct questions. The record identifies baseline and 12-month follow-up as NPS assessment points but does not state which approach was used. It also does not report whether all participants with any measurements were included, how incomplete records were handled, or whether the analysis followed a prespecified population.

What is over-dispersion, and is it relevant?

Over-dispersion is a general concern for some count-data models when observed variability exceeds what the model assumes. It is not directly indicated by the listed primary endpoint, which is a continuous patient-reported score summarized by a mean and standard deviation. The record does not identify count outcomes as the primary endpoint or describe models for secondary outcomes. It would therefore be inappropriate to infer that an over-dispersion adjustment was needed or used in this analysis.

How should the planning illustration be read?

The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect, and that non-time-to-event planning outputs are participants per arm for 80% or 90% power. This is a planning exercise, not evidence about the power achieved in the completed study. It is not a report of the trial's original sample-size justification. The record does not report the effect values or planning outputs needed to interpret a particular illustrated sample size, and it does not state that those assumptions governed the study design. Such illustrations should not be called observed power; post-study calculations based on an estimated effect do not establish the probability that the completed study detected a true effect.

In summary, statistical terminology describes what could be learned from means, variability, intervals and tests, but the record lacks the estimates and analysis details needed to apply those concepts to the primary comparison. The design and endpoint are documented more clearly than the statistical results.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and identifies NPS as the primary endpoint; ODI is named as a secondary outcome. The brief summary also describes pain, physical function, depression, pain medicine use, sleep, activity, healthcare resource utilization and cost as outcomes of interest. The registry record does not provide a complete mapping of these measures to outcome definitions, analysis time points or statistical comparisons. It does not report how the multiplicity of outcomes or comparisons was handled.

Multiplicity is a general statistical issue that arises when several hypotheses are tested. As the number of tests increases, the chance of at least one apparently unusual result can increase if each test is considered separately. Common approaches include prespecifying a primary hypothesis, adjusting significance thresholds, or describing secondary analyses as exploratory. The registry record does not state which approach, if any, was used. It would be inappropriate to infer a correction or a hierarchy of tests from the outcome list alone.

Missing-data handling is also not reported. Although the primary endpoint is scheduled at baseline and 12-month follow-up, the record does not state how many participants contributed measurements at either time, why observations might be missing, or whether any imputation or model-based method was used. General statistical concern: if availability of follow-up scores is related to pain or other participant characteristics, an analysis limited to observed scores may not represent the full group. Without missingness details and methods, the direction or magnitude of any such influence cannot be assessed.

The analysis populations are not described. The record does not say whether results were based on all enrolled participants, those who initiated WHT, those who completed follow-up, or another defined group. It also does not report protocol deviations or exclusions. The analysis count is 0, and no statistical plan is listed. Thus, neither the population used for efficacy analyses nor procedures addressing multiple outcomes and incomplete data are reported in the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The adverse-event entries report zero serious adverse events, zero other adverse events and zero deaths for the WHT group, the historic MPP control and the concurrent MPM control. The corresponding numbers at risk are 113 for WHT, 146 for the historic control and 161 for the concurrent control. These are the registry's reported counts for the listed groups.

These entries should be interpreted as recorded event counts, not as a complete account of safety experience. The registry record does not define serious or other adverse events, describe how events were collected, state the observation period for these safety summaries, or explain whether reporting was complete. It also does not provide event-level descriptions. A zero count in a registry entry does not establish that no event occurred; it means that no events in that category are reported in that entry.

The groups differ in care setting and exposure, and the record does not describe a randomized safety comparison. The counts therefore do not support a comparative safety estimate. They should not be treated as evidence that the groups had equivalent safety outcomes, nor do they quantify a difference in risk. The figures are best read narrowly as the reported serious-event, other-event and death counts alongside the listed numbers at risk.

12. What the Result Does, and Does Not, Mean

This registry record is useful for understanding the study's intended question, the wearable technology package, the care settings represented in the groups and the primary endpoint. It identifies a completed proof-of-concept study of WHT among patients treated for chronic pain at Geisinger. It also describes the NPS scale and its planned assessment at baseline and 12-month follow-up.

It does not, however, report the numerical primary-endpoint results. The group means, standard deviations, changes and between-group comparisons are not listed. There is no reported confidence interval or p-value for the primary outcome, and no analysis method is described. The effect's direction, magnitude and precision therefore cannot be determined from the registry record. The absence of a reported result is not evidence that the result was favorable, unfavorable or unchanged.

The record also does not establish a causal effect of wearable technology. The allocation method is not stated, and the record explicitly describes the study as nonrandomized with arms that are not like-for-like. The WHT group received MPP care, the historic control received MPP education without WHT, and the concurrent MPM control received neither MPP education nor WHT. General interpretation: differences across such groups could reflect their care programs, time periods or participant characteristics, in addition to any association with technology use. No adjustment strategy is reported.

The planning illustration described in the record should not be mistaken for observed study performance or an original design justification. It is framed as a calculation based on the observed effect and stated assumptions; it is not observed power. Without the underlying results or illustration outputs, it cannot supply an estimate of benefit or precision.

In short, the record documents a study framework and endpoint, but not enough statistical results to judge what the primary outcome showed. It does not establish that WHT improved pain, failed to improve pain, or had no effect. It also does not provide a detailed account of missing data, multiplicity, analysis populations or safety ascertainment.

13. Why This Trial Matters Statistically

This study raises a practical statistical question about evaluating a technology package within real care pathways. The intervention description combines a wearable device, phone, pain application and provider dashboards. The intended data stream includes self-reported pain and medication information as well as passively collected activity and sleep measures. Such a package can generate multiple types of outcomes, but the registry record does not detail how each measure was analyzed or prioritized beyond identifying NPS as primary and ODI as secondary.

The three-group structure is statistically important. The historic MPP control shares the educational program but differs in time period and lacks WHT. The concurrent MPM control shares the study period but differs in care program and does not receive MPP education. These comparisons could address different questions, but neither isolates technology exposure from all other differences based on the information reported. The record itself flags that the arms are not a like-for-like control comparison. This makes transparent design description and careful uncertainty reporting especially important.

Another important issue is the distinction between feasibility or proof-of-concept intent and evidence of an outcome effect. The record describes recruitment and follow-up intentions and names several clinical and utilization outcomes, but it does not report primary endpoint estimates. A study can be informative about a research approach even when the registry record does not contain a result that permits an effect estimate. Readers still need the numerical outcomes, analysis population, missing-data information and methods to judge the strength of evidence.

The registry also illustrates why a sample-size calculation based on an observed effect must be distinguished from a prospective design justification. The stated planning assumptions can support an illustrative calculation, but they do not show that the completed study had a particular probability of detecting an effect. With no primary outcome estimate or planning output listed, no sample-size illustration can resolve the uncertainty about the trial's findings.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific limitations follow from the information in the registry record. First, the allocation method is not stated, and the record explicitly characterizes the comparison as nonrandomized and not like-for-like. The three groups differ by care program, exposure to MPP education, or enrollment period. These design features complicate attribution of any outcome difference to WHT alone.

Second, the record does not report numerical results for the primary NPS endpoint. It provides the endpoint definition and assessment times but not the group summaries or comparative estimates. There is consequently no basis in this record for assessing effect size or precision. The count of six outcome measures and the named secondary outcome do not substitute for outcome-level results.

Third, key analysis details are absent. The registry record does not describe the statistical model, analysis population, handling of repeated measurements, missing-data method, multiplicity approach, or whether any adjustment was made for differences among groups. It does not report an interim-analysis plan or a sample-size justification for the study. These features are not to be inferred from a planning illustration described as based on an observed effect.

Finally, the safety entries lack definitions, collection procedures and ascertainment details. The listed zero counts are informative only as the registry's reported entries. They cannot establish complete event capture or comparative safety. The brief summary names additional outcomes and device-use information, but the record does not give corresponding result details. Together, these omissions mean that the registry record provides more information about the intended study question than about the study's analyzed findings.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effect of Wearable Health Technology on Patients Treated for Chronic Pain at Geisinger Health System — NCT03299556.
• Registry record last updated 2024-09-19; status last verified 2024-04.
• Results first posted 2024-09-19.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT03299556 is a completed proof-of-concept study sponsored by Purdue Pharma LP, with a listed enrollment of 420. It concerns chronic pain care at Geisinger and evaluates a wearable health technology package described as including an Apple Watch, iPhone, Pain App and provider dashboards.

The record describes a WHT group of newly enrolled MPP patients, a historic MPP control and a concurrent MPM control. Allocation is not stated; the record flags that this is not a randomized, like-for-like comparison. The primary outcome is mean NPS, with standard deviation, at baseline and 12-month follow-up. ODI is named as a secondary outcome.

Numerical results for the primary endpoint and details of the statistical analysis are not reported in the registry record. The listed adverse-event entries report zero serious events, zero other events and zero deaths in each group. These entries lack further ascertainment details.